[linux-2.6-block.git] / arch / s390 / mm / init.c

/*
 *  arch/s390/mm/init.c
 *
 *  S390 version
 *    Copyright (C) 1999 IBM Deutschland Entwicklung GmbH, IBM Corporation
 *    Author(s): Hartmut Penner (hp@de.ibm.com)
 *
 *  Derived from "arch/i386/mm/init.c"
 *    Copyright (C) 1995  Linus Torvalds
 */

#include <linux/signal.h>
#include <linux/sched.h>
#include <linux/kernel.h>
#include <linux/errno.h>
#include <linux/string.h>
#include <linux/types.h>
#include <linux/ptrace.h>
#include <linux/mman.h>
#include <linux/mm.h>
#include <linux/swap.h>
#include <linux/smp.h>
#include <linux/init.h>
#include <linux/pagemap.h>
#include <linux/bootmem.h>
#include <linux/pfn.h>
#include <linux/poison.h>
#include <linux/initrd.h>
#include <linux/export.h>
#include <linux/gfp.h>
#include <asm/processor.h>
#include <asm/uaccess.h>
#include <asm/pgtable.h>
#include <asm/pgalloc.h>
#include <asm/dma.h>
#include <asm/lowcore.h>
#include <asm/tlb.h>
#include <asm/tlbflush.h>
#include <asm/sections.h>
#include <asm/ctl_reg.h>

pgd_t swapper_pg_dir[PTRS_PER_PGD] __attribute__((__aligned__(PAGE_SIZE)));

unsigned long empty_zero_page, zero_page_mask;
EXPORT_SYMBOL(empty_zero_page);

static unsigned long setup_zero_pages(void)
{
	struct cpuid cpu_id;
	unsigned int order;
	unsigned long size;
	struct page *page;
	int i;

	get_cpu_id(&cpu_id);
	switch (cpu_id.machine) {
	case 0x9672:	/* g5 */
	case 0x2064:	/* z900 */
	case 0x2066:	/* z900 */
	case 0x2084:	/* z990 */
	case 0x2086:	/* z990 */
	case 0x2094:	/* z9-109 */
	case 0x2096:	/* z9-109 */
		order = 0;
		break;
	case 0x2097:	/* z10 */
	case 0x2098:	/* z10 */
	default:
		order = 2;
		break;
	}

	empty_zero_page = __get_free_pages(GFP_KERNEL | __GFP_ZERO, order);
	if (!empty_zero_page)
		panic("Out of memory in setup_zero_pages");

	page = virt_to_page((void *) empty_zero_page);
	split_page(page, order);
	for (i = 1 << order; i > 0; i--) {
		SetPageReserved(page);
		page++;
	}

	size = PAGE_SIZE << order;
	zero_page_mask = (size - 1) & PAGE_MASK;

	return 1UL << order;
}

/*
 * paging_init() sets up the page tables
 */
void __init paging_init(void)
{
	unsigned long max_zone_pfns[MAX_NR_ZONES];
	unsigned long pgd_type, asce_bits;

	init_mm.pgd = swapper_pg_dir;
#ifdef CONFIG_64BIT
	if (VMALLOC_END > (1UL << 42)) {
		asce_bits = _ASCE_TYPE_REGION2 | _ASCE_TABLE_LENGTH;
		pgd_type = _REGION2_ENTRY_EMPTY;
	} else {
		asce_bits = _ASCE_TYPE_REGION3 | _ASCE_TABLE_LENGTH;
		pgd_type = _REGION3_ENTRY_EMPTY;
	}
#else
	asce_bits = _ASCE_TABLE_LENGTH;
	pgd_type = _SEGMENT_ENTRY_EMPTY;
#endif
	S390_lowcore.kernel_asce = (__pa(init_mm.pgd) & PAGE_MASK) | asce_bits;
	clear_table((unsigned long *) init_mm.pgd, pgd_type,
		    sizeof(unsigned long)*2048);
	vmem_map_init();

        /* enable virtual mapping in kernel mode */
	__ctl_load(S390_lowcore.kernel_asce, 1, 1);
	__ctl_load(S390_lowcore.kernel_asce, 7, 7);
	__ctl_load(S390_lowcore.kernel_asce, 13, 13);
	arch_local_irq_restore(4UL << (BITS_PER_LONG - 8));

	atomic_set(&init_mm.context.attach_count, 1);

	sparse_memory_present_with_active_regions(MAX_NUMNODES);
	sparse_init();
	memset(max_zone_pfns, 0, sizeof(max_zone_pfns));
	max_zone_pfns[ZONE_DMA] = PFN_DOWN(MAX_DMA_ADDRESS);
	max_zone_pfns[ZONE_NORMAL] = max_low_pfn;
	free_area_init_nodes(max_zone_pfns);
	fault_init();
}

void __init mem_init(void)
{
	unsigned long codesize, reservedpages, datasize, initsize;

        max_mapnr = num_physpages = max_low_pfn;
        high_memory = (void *) __va(max_low_pfn * PAGE_SIZE);

	/* Setup guest page hinting */
	cmma_init();

	/* this will put all low memory onto the freelists */
	totalram_pages += free_all_bootmem();
	totalram_pages -= setup_zero_pages();	/* Setup zeroed pages. */

	reservedpages = 0;

	codesize =  (unsigned long) &_etext - (unsigned long) &_text;
	datasize =  (unsigned long) &_edata - (unsigned long) &_etext;
	initsize =  (unsigned long) &__init_end - (unsigned long) &__init_begin;
        printk("Memory: %luk/%luk available (%ldk kernel code, %ldk reserved, %ldk data, %ldk init)\n",
		nr_free_pages() << (PAGE_SHIFT-10),
                max_mapnr << (PAGE_SHIFT-10),
                codesize >> 10,
                reservedpages << (PAGE_SHIFT-10),
                datasize >>10,
                initsize >> 10);
	printk("Write protected kernel read-only data: %#lx - %#lx\n",
	       (unsigned long)&_stext,
	       PFN_ALIGN((unsigned long)&_eshared) - 1);
}

#ifdef CONFIG_DEBUG_PAGEALLOC
void kernel_map_pages(struct page *page, int numpages, int enable)
{
	pgd_t *pgd;
	pud_t *pud;
	pmd_t *pmd;
	pte_t *pte;
	unsigned long address;
	int i;

	for (i = 0; i < numpages; i++) {
		address = page_to_phys(page + i);
		pgd = pgd_offset_k(address);
		pud = pud_offset(pgd, address);
		pmd = pmd_offset(pud, address);
		pte = pte_offset_kernel(pmd, address);
		if (!enable) {
			__ptep_ipte(address, pte);
			pte_val(*pte) = _PAGE_TYPE_EMPTY;
			continue;
		}
		*pte = mk_pte_phys(address, __pgprot(_PAGE_TYPE_RW));
		/* Flush cpu write queue. */
		mb();
	}
}
#endif

void free_init_pages(char *what, unsigned long begin, unsigned long end)
{
	unsigned long addr = begin;

	if (begin >= end)
		return;
	for (; addr < end; addr += PAGE_SIZE) {
		ClearPageReserved(virt_to_page(addr));
		init_page_count(virt_to_page(addr));
		memset((void *)(addr & PAGE_MASK), POISON_FREE_INITMEM,
		       PAGE_SIZE);
		free_page(addr);
		totalram_pages++;
	}
	printk(KERN_INFO "Freeing %s: %luk freed\n", what, (end - begin) >> 10);
}

void free_initmem(void)
{
	free_init_pages("unused kernel memory",
			(unsigned long)&__init_begin,
			(unsigned long)&__init_end);
}

#ifdef CONFIG_BLK_DEV_INITRD
void free_initrd_mem(unsigned long start, unsigned long end)
{
	free_init_pages("initrd memory", start, end);
}
#endif

#ifdef CONFIG_MEMORY_HOTPLUG
int arch_add_memory(int nid, u64 start, u64 size)
{
	unsigned long zone_start_pfn, zone_end_pfn, nr_pages;
	unsigned long start_pfn = PFN_DOWN(start);
	unsigned long size_pages = PFN_DOWN(size);
	struct zone *zone;
	int rc;

	rc = vmem_add_mapping(start, size);
	if (rc)
		return rc;
	for_each_zone(zone) {
		if (zone_idx(zone) != ZONE_MOVABLE) {
			/* Add range within existing zone limits */
			zone_start_pfn = zone->zone_start_pfn;
			zone_end_pfn = zone->zone_start_pfn +
				       zone->spanned_pages;
		} else {
			/* Add remaining range to ZONE_MOVABLE */
			zone_start_pfn = start_pfn;
			zone_end_pfn = start_pfn + size_pages;
		}
		if (start_pfn < zone_start_pfn || start_pfn >= zone_end_pfn)
			continue;
		nr_pages = (start_pfn + size_pages > zone_end_pfn) ?
			   zone_end_pfn - start_pfn : size_pages;
		rc = __add_pages(nid, zone, start_pfn, nr_pages);
		if (rc)
			break;
		start_pfn += nr_pages;
		size_pages -= nr_pages;
		if (!size_pages)
			break;
	}
	if (rc)
		vmem_remove_mapping(start, size);
	return rc;
}
#endif /* CONFIG_MEMORY_HOTPLUG */
Commit	Line	Data
1da177e4 LT	1	/*
	2	* arch/s390/mm/init.c
	3	*
	4	* S390 version
	5	* Copyright (C) 1999 IBM Deutschland Entwicklung GmbH, IBM Corporation
	6	* Author(s): Hartmut Penner (hp@de.ibm.com)
	7	*
	8	* Derived from "arch/i386/mm/init.c"
	9	* Copyright (C) 1995 Linus Torvalds
	10	*/
	11
1da177e4 LT	12	#include <linux/signal.h>
	13	#include <linux/sched.h>
	14	#include <linux/kernel.h>
	15	#include <linux/errno.h>
	16	#include <linux/string.h>
	17	#include <linux/types.h>
	18	#include <linux/ptrace.h>
	19	#include <linux/mman.h>
	20	#include <linux/mm.h>
	21	#include <linux/swap.h>
	22	#include <linux/smp.h>
	23	#include <linux/init.h>
	24	#include <linux/pagemap.h>
	25	#include <linux/bootmem.h>
d882b172	26	#include <linux/pfn.h>
028d9b3c	27	#include <linux/poison.h>
2b67fc46	28	#include <linux/initrd.h>
3a4c5d59	29	#include <linux/export.h>
5a0e3ad6	30	#include <linux/gfp.h>
1da177e4	31	#include <asm/processor.h>
1da177e4 LT	32	#include <asm/uaccess.h>
	33	#include <asm/pgtable.h>
	34	#include <asm/pgalloc.h>
	35	#include <asm/dma.h>
	36	#include <asm/lowcore.h>
	37	#include <asm/tlb.h>
	38	#include <asm/tlbflush.h>
d882b172	39	#include <asm/sections.h>
a0616cde	40	#include <asm/ctl_reg.h>
1da177e4	41
1da177e4	42	pgd_t swapper_pg_dir[PTRS_PER_PGD] __attribute__((__aligned__(PAGE_SIZE)));
1485c5c8	43
238ec4ef	44	unsigned long empty_zero_page, zero_page_mask;
1485c5c8	45	EXPORT_SYMBOL(empty_zero_page);
1da177e4	46
238ec4ef MS	47	static unsigned long setup_zero_pages(void)
	48	{
	49	struct cpuid cpu_id;
	50	unsigned int order;
	51	unsigned long size;
	52	struct page *page;
	53	int i;
	54
	55	get_cpu_id(&cpu_id);
	56	switch (cpu_id.machine) {
	57	case 0x9672: /* g5 */
	58	case 0x2064: /* z900 */
	59	case 0x2066: /* z900 */
	60	case 0x2084: /* z990 */
	61	case 0x2086: /* z990 */
	62	case 0x2094: /* z9-109 */
	63	case 0x2096: /* z9-109 */
	64	order = 0;
	65	break;
	66	case 0x2097: /* z10 */
	67	case 0x2098: /* z10 */
	68	default:
	69	order = 2;
	70	break;
	71	}
	72
	73	empty_zero_page = __get_free_pages(GFP_KERNEL \| __GFP_ZERO, order);
	74	if (!empty_zero_page)
	75	panic("Out of memory in setup_zero_pages");
	76
	77	page = virt_to_page((void *) empty_zero_page);
	78	split_page(page, order);
	79	for (i = 1 << order; i > 0; i--) {
	80	SetPageReserved(page);
	81	page++;
	82	}
	83
	84	size = PAGE_SIZE << order;
	85	zero_page_mask = (size - 1) & PAGE_MASK;
	86
	87	return 1UL << order;
	88	}
	89
1da177e4 LT	90	/*
	91	* paging_init() sets up the page tables
	92	*/
1da177e4 LT	93	void __init paging_init(void)
1da177e4 LT	94	{
39b742f9	95	unsigned long max_zone_pfns[MAX_NR_ZONES];
14045ebf	96	unsigned long pgd_type, asce_bits;
d882b172	97
3610cce8	98	init_mm.pgd = swapper_pg_dir;
f4eb07c1	99	#ifdef CONFIG_64BIT
14045ebf MS	100	if (VMALLOC_END > (1UL << 42)) {
	101	asce_bits = _ASCE_TYPE_REGION2 \| _ASCE_TABLE_LENGTH;
	102	pgd_type = _REGION2_ENTRY_EMPTY;
	103	} else {
	104	asce_bits = _ASCE_TYPE_REGION3 \| _ASCE_TABLE_LENGTH;
	105	pgd_type = _REGION3_ENTRY_EMPTY;
	106	}
f4eb07c1	107	#else
14045ebf	108	asce_bits = _ASCE_TABLE_LENGTH;
3610cce8	109	pgd_type = _SEGMENT_ENTRY_EMPTY;
f4eb07c1	110	#endif
14045ebf	111	S390_lowcore.kernel_asce = (__pa(init_mm.pgd) & PAGE_MASK) \| asce_bits;
3610cce8 MS	112	clear_table((unsigned long *) init_mm.pgd, pgd_type,
3610cce8 MS	113	sizeof(unsigned long)*2048);
f4eb07c1	114	vmem_map_init();
1da177e4	115
1da177e4	116	/* enable virtual mapping in kernel mode */
3610cce8 MS	117	__ctl_load(S390_lowcore.kernel_asce, 1, 1);
	118	__ctl_load(S390_lowcore.kernel_asce, 7, 7);
	119	__ctl_load(S390_lowcore.kernel_asce, 13, 13);
df9ee292	120	arch_local_irq_restore(4UL << (BITS_PER_LONG - 8));
1da177e4	121
050eef36 MS	122	atomic_set(&init_mm.context.attach_count, 1);
050eef36 MS	123
17f34580 HC	124	sparse_memory_present_with_active_regions(MAX_NUMNODES);
17f34580 HC	125	sparse_init();
39b742f9 HC	126	memset(max_zone_pfns, 0, sizeof(max_zone_pfns));
	127	max_zone_pfns[ZONE_DMA] = PFN_DOWN(MAX_DMA_ADDRESS);
	128	max_zone_pfns[ZONE_NORMAL] = max_low_pfn;
	129	free_area_init_nodes(max_zone_pfns);
92f842ea	130	fault_init();
1da177e4	131	}
1da177e4 LT	132
	133	void __init mem_init(void)
	134	{
	135	unsigned long codesize, reservedpages, datasize, initsize;
	136
	137	max_mapnr = num_physpages = max_low_pfn;
	138	high_memory = (void ) __va(max_low_pfn PAGE_SIZE);
	139
45e576b1 MS	140	/* Setup guest page hinting */
	141	cmma_init();
	142
1da177e4 LT	143	/* this will put all low memory onto the freelists */
1da177e4 LT	144	totalram_pages += free_all_bootmem();
238ec4ef	145	totalram_pages -= setup_zero_pages(); /* Setup zeroed pages. */
1da177e4 LT	146
	147	reservedpages = 0;
	148
	149	codesize = (unsigned long) &_etext - (unsigned long) &_text;
	150	datasize = (unsigned long) &_edata - (unsigned long) &_etext;
	151	initsize = (unsigned long) &__init_end - (unsigned long) &__init_begin;
	152	printk("Memory: %luk/%luk available (%ldk kernel code, %ldk reserved, %ldk data, %ldk init)\n",
cc013a88	153	nr_free_pages() << (PAGE_SHIFT-10),
1da177e4 LT	154	max_mapnr << (PAGE_SHIFT-10),
	155	codesize >> 10,
	156	reservedpages << (PAGE_SHIFT-10),
	157	datasize >>10,
	158	initsize >> 10);
d882b172	159	printk("Write protected kernel read-only data: %#lx - %#lx\n",
162e006e HC	160	(unsigned long)&_stext,
162e006e HC	161	PFN_ALIGN((unsigned long)&_eshared) - 1);
1da177e4 LT	162	}
1da177e4 LT	163
2485579b HC	164	#ifdef CONFIG_DEBUG_PAGEALLOC
	165	void kernel_map_pages(struct page *page, int numpages, int enable)
	166	{
	167	pgd_t *pgd;
	168	pud_t *pud;
	169	pmd_t *pmd;
	170	pte_t *pte;
	171	unsigned long address;
	172	int i;
	173
	174	for (i = 0; i < numpages; i++) {
	175	address = page_to_phys(page + i);
	176	pgd = pgd_offset_k(address);
	177	pud = pud_offset(pgd, address);
	178	pmd = pmd_offset(pud, address);
	179	pte = pte_offset_kernel(pmd, address);
	180	if (!enable) {
b2fa47e6 MS	181	__ptep_ipte(address, pte);
b2fa47e6 MS	182	pte_val(*pte) = _PAGE_TYPE_EMPTY;
2485579b HC	183	continue;
	184	}
	185	*pte = mk_pte_phys(address, __pgprot(_PAGE_TYPE_RW));
	186	/* Flush cpu write queue. */
	187	mb();
	188	}
	189	}
	190	#endif
	191
d96221ab	192	void free_init_pages(char *what, unsigned long begin, unsigned long end)
1da177e4	193	{
d96221ab	194	unsigned long addr = begin;
1da177e4	195
d96221ab HC	196	if (begin >= end)
	197	return;
	198	for (; addr < end; addr += PAGE_SIZE) {
1da177e4	199	ClearPageReserved(virt_to_page(addr));
7835e98b	200	init_page_count(virt_to_page(addr));
d96221ab HC	201	memset((void *)(addr & PAGE_MASK), POISON_FREE_INITMEM,
d96221ab HC	202	PAGE_SIZE);
1da177e4 LT	203	free_page(addr);
1da177e4 LT	204	totalram_pages++;
d96221ab HC	205	}
	206	printk(KERN_INFO "Freeing %s: %luk freed\n", what, (end - begin) >> 10);
	207	}
	208
	209	void free_initmem(void)
	210	{
	211	free_init_pages("unused kernel memory",
	212	(unsigned long)&__init_begin,
	213	(unsigned long)&__init_end);
1da177e4 LT	214	}
	215
	216	#ifdef CONFIG_BLK_DEV_INITRD
	217	void free_initrd_mem(unsigned long start, unsigned long end)
	218	{
d96221ab	219	free_init_pages("initrd memory", start, end);
1da177e4 LT	220	}
1da177e4 LT	221	#endif
421c175c HC	222
	223	#ifdef CONFIG_MEMORY_HOTPLUG
	224	int arch_add_memory(int nid, u64 start, u64 size)
	225	{
892365ab GS	226	unsigned long zone_start_pfn, zone_end_pfn, nr_pages;
	227	unsigned long start_pfn = PFN_DOWN(start);
	228	unsigned long size_pages = PFN_DOWN(size);
421c175c HC	229	struct zone *zone;
	230	int rc;
	231
421c175c HC	232	rc = vmem_add_mapping(start, size);
	233	if (rc)
	234	return rc;
892365ab GS	235	for_each_zone(zone) {
	236	if (zone_idx(zone) != ZONE_MOVABLE) {
	237	/* Add range within existing zone limits */
	238	zone_start_pfn = zone->zone_start_pfn;
	239	zone_end_pfn = zone->zone_start_pfn +
	240	zone->spanned_pages;
	241	} else {
	242	/* Add remaining range to ZONE_MOVABLE */
	243	zone_start_pfn = start_pfn;
	244	zone_end_pfn = start_pfn + size_pages;
	245	}
	246	if (start_pfn < zone_start_pfn \|\| start_pfn >= zone_end_pfn)
	247	continue;
	248	nr_pages = (start_pfn + size_pages > zone_end_pfn) ?
	249	zone_end_pfn - start_pfn : size_pages;
	250	rc = __add_pages(nid, zone, start_pfn, nr_pages);
	251	if (rc)
	252	break;
	253	start_pfn += nr_pages;
	254	size_pages -= nr_pages;
	255	if (!size_pages)
	256	break;
	257	}
421c175c HC	258	if (rc)
	259	vmem_remove_mapping(start, size);
	260	return rc;
	261	}
	262	#endif /* CONFIG_MEMORY_HOTPLUG */